Model Terminal

Qwen Fixed Chat Templates

Drop-in Jinja chat templates for Qwen 3.5 and 3.6 that fix known bugs in the official templates, improving tool-calling, reasoning-tag handling, and compatibility with local inference engines such as LM Studio, llama.cpp, MLX, and vLLM. Not a model in the weight sense — a prompt-formatting utility published on Hugging Face. Source: supabase.

Value score
Not scored
Context
tokens
Max output
tokens
Price

Capability radar

Identity

Developer
froggeric
Openness
Modalities
Release
Knowledge cutoff
Deprecation
API docs

Benchmarks

No benchmark scores yet.

Compare nearby

Pricing

Input / 1M
Output / 1M
Speed